SayMotion is a browser-based generative AI service for creating character animations from textual prompts. Founded in 2014 by gaming industry veteran Kevin He, DeepMotion is a pioneer in motion capture using AI and has paved the way for recent browser-based mocap tools. With SayMotion, the company has ventured into the realm of generative AI, allowing users to create animated sequences by inputting textual descriptions of movements they wish to generate. Descriptions are categorized into three subsections: theme, action, and details.

In the SayMotion 1.5 update, DeepMotion has expanded the size of the dataset used to train the SayMotion AI model to 400,000 animation clips. Additionally, new movement categories, including parkour, have been introduced. The drawing system has been enhanced to allow users to insert an animation clip at the beginning of a drawing session and trim the beginning or end of clips to remove unwanted parts of the motion.
Currently in open beta, SayMotion is compatible with any standard desktop web browser, with plans for mobile browser support in the future. During the beta testing period, users receive 200 credits per month. Each credit is consumed when creating an animation or MP4, uploading animation, or using inpainting features. Animations created using beta accounts are licensed for commercial use. DeepMotion has not yet announced the release date or final pricing for SayMotion.